So now you can fall in love with someone that doesn't exist.
Comment has been collapsed.
Been doing that since I discovered anime. D: Now I got 3d waifu's too? Shieeeeeeeeeeeet.
Comment has been collapsed.
This is fascinating, love how AI is improving each day!
Comment has been collapsed.
This is kind of cool and to be honest there's far more scary things to be dealt with in the future.
Comment has been collapsed.
You think if i click enough times for the next 5 years i will find one that looks like me
Comment has been collapsed.
Looking at this I can only think catfishers, bot accounts and the like will now have the perfect source of untraceable, original people pics to use... Not that any of them are too attractive looking from what I've clicked through. Most of them do still look rather unnatural, especially when you focus on their skin or teeth. They sorta look like those monsters that wear stolen human skin... And they're always looking directly at you with that soulless stare and empty smile... Yeah, definitely creepy. Undeniably good tech though, but like all advancement of this kind--scary.
Some of the creepiest/weirdest faces I've found:
https://i.imgur.com/hcPxmrF.jpg
the stare.. the smile.. WHAT IS SHE ((THEY)) PLANNING?
https://i.imgur.com/cwcPvUO.jpg
SCP looking mother fucker on the left
https://i.imgur.com/rYeruT0.jpg
some sort of Junji Ito ass hole on this boy's forehead
https://i.imgur.com/8MCmFPI.jpg
dude's face is melting and spawning new faces?? the closer you look the more gross it is. I swear you can see blood and stringy wet bits
https://i.imgur.com/JCoSWla.jpg
probably one of the "best" looking faces I've seen if you can ignore the glitches
https://i.imgur.com/LTzAc5o.jpg
DISGUISE. ERROR. DISGUISE. ERROR.
Someone could definitely write a good SCP entry about this. Maybe involving you eventually finding your own exact face if you click enough times, staring--like all the rest--straight back at you, at an angle and place you know you've never taken or been before, and then... "It" captures you. Or something. Maybe you find a dead relative... You get the idea.
Comment has been collapsed.
I tested 3 pictures.
1x creepy
1x ok
1x good
Looks like as a normal female use a filter for her picture and choosen a bad picture quality from converting or something.
Without the knowledge that it is complete fake i don't seen it.
Comment has been collapsed.
All I can think about is how dope this would be as a character generator for stories. I'm gonna save this link, too, when I want to practice portraits. I've refreshed a bunch of times but haven't gotten any weird ones.
Comment has been collapsed.
Hmm. Great site to fill data on fake facebook accounts.
Comment has been collapsed.
Combine with Fake Person Generator and you've got yourself a throwaway profile.
Comment has been collapsed.
Luckily those people always have signs that they're fake. Rendering glitches, issues with hair, blending and so on.
I've been on this site multiple times though. Pretty weird stuff,
Comment has been collapsed.
The point between the left ear and head has rendering glitches. The point between the forehead and the spot above the eyes has inconsistent skin color. Wrinkles, like Marcio pointed out, are inconsistent and don't fit her age either. Look at her right shoulder and check out her hair. Low polygon counts and it's obvious there's some distortion there.
They seem to split the head up into multiple parts. Ears, forehead, eyes, nose, chin, etc.
But obviously, this thing's insane. If you weren't looking out for these errors, then odds are it'd fool most people, including me.
Comment has been collapsed.
Interesting to see what computers and AI can accomplish nowadays. Unfortunately, even with this, for whatever reason computers still cannot reproduce the human eye without making them eternally cross eyed.
Comment has been collapsed.
Looks like the deception does not work on partial faces...
Comment has been collapsed.
I've just looked at a couple of dozen and aside from the glitches and artefacts that other people have noticed the facial structure in none of the faces looked right. I possibly look at people very differently now due to the amount of portraits I draw/paint but none of those looked like good fakes to me.
Comment has been collapsed.
Pretty sure there´s something from Charlize Theron in there:
Comment has been collapsed.
A one-off website demonstrated the power of new Nvidia developed neural networks with addictive and creepy results.
Open the website www.thispersondoesnotexist.com and you’ll find a headshot of a stranger. At first glance it looks like someone got the HR records from work and stuck them on a website, refreshing the site regenerates another face of a person you might know.
Except you don’t, and no one does. Because each of the faces on this site is totally fake, created by using a special kind of artificial intelligence algorithm called generative adversarial networks (GANs).
Refresh to the page and the algorithm generates a hyper-realistic but absolutely fake image of a person. The page was set up by Uber software engineer Philip Wang to demonstrate what GANs are capable of. Wang shared the page on the Facebook group “Artificial Intelligence & Deep Learning”.
The code that made this creepy website possible was written by Nvidia and featured in a paper that is available before peer review on arXiv. Called StyleGAN, the neural network has infinite applicability for everything from gaming to creating false documents.
Wang described his motivation for sharing the site in his Facebook post: “I have decided to dig into my own pockets and raise some public awareness for this technology,” he wrote in his post.
“Faces are most salient to our cognition, so I’ve decided to put that specific pre-trained model up. Each time you refresh the site, the network will generate a new facial image from scratch from a 512-dimensional vector.”
How it works:
All GAN’s have two networks: the generator and the discriminator. The generator synthesizes new samples from scratch, and the discriminator takes samples from both the training data and the generator’s output and predicts if they are “real” or “fake”.
The generator receives a random vector (noise) and therefore its initial output is also noise.
After it receives feedback from the discriminator, it learns to synthesize more “realistic” images. Simultaneously the discriminator is also learning by comparing generated samples with real samples, making it harder for the generator to deceive it.
GAN was introduced in 2014 but it wasn’t until 2017 that the researchers were able to create high-quality, 1024x1024 images detailed in the now-famous ProGAN paper. StyleGAN builds on this previous work but now allows researchers more control over specific features.
GAN's will shape the virtual future
Eventually, these GAN’s are hoped to be able to be used to develop full virtual worlds using automated methods instead of hard coding. They also have the possibility of being used to create realistic 3D models for use in advertising and other branding opportunities, similar to Imma but with a little more depth.
While that is exciting, others may fear for the more sinister uses for the technology such as contributing to DeepFakes, computer-generated images superimposed on existing pictures or videos, that can be used to push fake news. Some thoughts to ponder on while you constantly refreshing this person does not exist.
Comment has been collapsed.